372 SSSAJ: Volume 71: Number 2 • March –April 2007

Soil Sci. Soc. Am. J. 71:372–379doi:10.2136/sssaj2006.0225Received 13 June 2006. *Corresponding author (laurie@maine.edu).© Soil Science Society of America677 S. Segoe Rd. Madison WI 53711 USA

Soils play an important role in the global C cycle in that they store large amounts of C, and if managed to do so, may provide an

enhanced sink for atmospheric CO2. Previous studies have shown that the organic matter in the world’s soils contain as much as three times the C found in land vegetation (Schlesinger, 1986; Raymond and Bauer, 2001). In addition, soil C has much longer residence mean times than the C in the vegetation that the soils support. Storage of organic C in this long-residence-time pool is referred to as carbon sequestration (Wang and Hseih, 2002).

The concern about increasing atmospheric CO2 and its role in future global climate change (Fung et al., 2005; Matthews et al., 2005) has led soil scientists to quantify soil C content (also referred to as storage or stocks) and residence times in specifi c locations to better constrain global C budgets. Identifi cation of age and turnover time is achieved using radiocarbon measurements (e.g., Knorr et al., 2005; Gaudinski et al., 2000). The common method of quantifying soil organic carbon (OC) is to use the total soil OC data for specifi c sites and scale up to regional or global estimates using soil maps (Lacelle et al., 2001) or soil-landscape modeling (Thompson and Kolka, 2005). Examples of such soil C accounting are the estimates of soil C stocks in Maine (Davidson and Lefebvre, 1993) and the conterminous USA (Guo et al., 2006).

The soils of swamps and marshes store the greatest amount of OC of all terrestrial ecosystems—up to 1000 Mg ha−1 in the top 1 m (Houghton, 1995; Trumbore and Harden, 1997; U.S. Department of Energy, 1999)—and tend to have very long

turnover times (Choi and Wang, 2004). Carbon accumulates in these soils because anaerobic conditions limit the rate of organic matter decomposition (Brady and Weil, 1996). In an attempt to improve soil C budgets, OC storage in other landscapes with anaerobic conditions have been quantifi ed. Examples are man-grove forests and tidal marshes (Chmura et al., 2003; Choi and Wang, 2004; Hussein et al., 2004), subsurface horizons of ripar-ian soils (Blazejewski et al., 2005), and reservoirs and other man-made impoundments (Smith et al., 2001).

Despite the focused efforts to balance the C budget, global change scientists have not been able to identify in what form some of the atmospheric CO2 is being stored annually (Pacala et al., 2001). Models of the C cycle identify the location of this unknown sink as somewhere in the midlatitude terrestrial bio-sphere (Houghton, 2003). Approximately half of this C is being stored in the aboveground biomass of North American forests (Hurtt et al., 2002; Pacala et al., 2001), but the location of the rest of the sink is unknown (Prentice et al., 2001). The challenge of fi nding the unknown C sink requires looking for other ecosys-tems in the region for which the aerial extent and soil C contents are, thus far, poorly quantifi ed. The non-marsh estuarine ecosys-tems along the midlatitude Atlantic Coast fi t into this category.

Estuaries have the highest primary productivity of all ecosys-tems (Odum, 1959). The allochthonous (photosynthesized in the estuary) and autochthonous (from outside the estuary) organic matter is processed within the estuary by fi lter feeders, decomposed and remineralized by benthic organisms, and moved through the system by tidal fl uctuations (Potsma, 1967). Seagrasses, the rooted plants in non-marsh subtidal estuaries, stabilize sediments, dampen wave action, and collect particulates, including organic matter, from the water column (Gacia and Duarte, 2001; Gacia et al., 1999). The particles intercepted by the grasses are then depos-ited on the soil surface. As additional sediment is deposited, the

Carbon Storage in the Soils of a Mesotidal Gulf of Maine Estuary

Estuaries have the highest primary productivity of all ecosystems, yet the soils in estuarine environments have been largely overlooked in C sequestration studies. This research quantifi ed organic C stored in the top 1 m of soils from the Taunton Bay estuary in Hancock County, Maine. Taunton Bay is centrally located within Maine’s Island–Bay Coastal Complex, a phys-iographic region that encompasses 47% of the state’s coastline. Riverine inputs of sediments and sediment-associated organic matter to this and most Maine estuaries are quite low. These estuary soils form by incremental additions of reworked glaciofl uvial marine sediments and estuarine organic matter to the soil surface. The organics are protected from decomposition by anaerobic conditions. As sea level rises, the estuary area increases by inundating and submerg-ing the soils at its edges. The average concentration of soil organic C in the top 1 m of these estuary soils is 2.4% and the average soil bulk density is 0.67 g cm−3. The organic C content is 136 Mg C ha−1, which is greater than the C content in the top 1 m of Maine’s upland soils. These results suggest that systematically quantifying and dating the C in estuarine soils will provide valuable data for use in regional and global C budgets and climate models.

J. L. JespersenDep. of Plant, Soil and Environ.Sci.Univ. of MaineOrono, ME 04469-5722

present address:FB Environmental1 India St.Portland, ME 04101

L. J. Osher*Dep. of Plant, Soil and Environmental Sci. Univ. of MaineOrono, ME 04469-5722

Abbreviations: BB, bay bottom landscape unit; CS, channel shoulder landscape unit; DEC, deep edge and cove landscape unit; FES, fi ne-silty and fi ne-loamy estuarine sediments; MS, mussel shoal landscape unit; OC, organic carbon; RSEC, recently submerged edge and cove landscape unit; SFSM, submerged fl uvial stream and marshes landscape unit.

PED

OLO

GY

Published online March 12, 2007

SSSAJ: Volume 71: Number 2 • March –April 2007 373

anoxic and light-limited conditions below the sediment surface inhibit microbial processing and photodegradation.

Despite these ideal conditions for production and preserva-tion of organic matter, the soils of coastal ecosystems, specifi cally estuarine systems, have been largely overlooked in C sequestration studies (Chmura et al., 2003; Thom et al., 2003). This is, in part, because soil maps are limited for submerged environments (Bradley and Stolt, 2003). A few detailed soil maps have been completed for some midlatitude Atlantic coastal environments in the USA (Bradley and Stolt, 2003; Demas and Rabenhorst, 1999; Demas et al., 1996; New York City Soil Survey Staff, 2005). Several other soil maps of estuarine environments are in progress (Balduff and Rabenhorst, 2006; Coppock and Rabenhorst, 2003; Fischler et al., 2005; Payne and Stolt, 2006; Saunders and Collins, 2006). The soil map for Taunton Bay, Maine, was recently completed (Flannagan and Osher, 2006). The maps generated and samples collected for that study were used as resources for this research.

This research quantifi es the total soil OC to a depth of 100 cm in the soils of Taunton Bay, a shallow, rock-bound, mesotidal estuary on the central Maine coast. This study is the fi rst accounting of C storage in non-marsh estuarine soils for use in regional soil C budgets and climate models. These data were used to address the hypothesis that Maine’s estuarine soils store more C per hectare than Maine’s upland soils.

STUDY SITETaunton Bay, a 1329-ha estuary, is located in Maine’s

Downeast region along the section of the coast characterized as the Island–Bay Complex (Jacobson et al., 1987; Fig. 1). This region includes the area from the western margin of Penobscot Bay to just north of Machias Bay, and encompasses 47% of Maine’s total coastline (Kelley, 1987). Estuaries in this portion of Maine’s north-central coast are glacially incised, have bedrock outcrops that form rounded granitic islands, and have extensive fi ne-grained mudfl ats (Jacobson et al., 1987; Larsen et al., 1980). In general, Maine’s estuaries have large tidal ranges (2–11 m) and small riverine sedi-ment supplies (Folger, 1972). The combined effects of low sedi-ment input and bedrock relief in this area limit marsh formation; the only marshes in this coastal region are fringing marshes at the edges of the most quiescent coves (Kelley, 1987).

The sediments being deposited in these estuaries are silts and silty clays that have eroded from coastal bluffs (Kelley, 1987). The bluffs are composed of the thick glaciomarine sediment layer known as the Presumpscot Formation (Bloom, 1960; Schnitker, 1974). This material, which was deposited between 14 000 and 11 500 yr BP, blankets portions of eastern Maine’s uplands as far as 100 km inland from the present coastline (Dorion et al., 2001).

Sea level has been rising consistently in Maine for the past 10 000 yr (Belknap et al., 1988), with a rapid increase of 30 to 40 cm in the last 200 yr (Gehrels et al., 2002). Sea-level rise causes wave erosion and destabilization of bluffs composed of glaciomarine parent materials (Kelley, 1987). Along the estuarine–terrestrial edges, unarmored and unvegetated soils formed in till and glaciomarine parent materials are easily dis-lodged, becoming the sediment source for the adjacent estuary (Anderson et al., 1981; Jordan, 1999). As sea level rises, the estuaries receive incremental additions of fi ne sediments and organic matter to the soil surface, and previously deposited material is buried (Belknap et al., 1988).

The Taunton Bay estuary is mostly shallow, with narrow channels as deep as 21 m below mean sea level. The average water depth at mean sea level is <3 m. Mean tidal range is 2.7 m and mean spring tidal range is 3.4 m (Flannagan and Osher, 2006). The streams entering the estuary include West Brook, Egypt Stream, and more than a dozen named and unnamed streams draining the western, northern, and southeastern por-tions of the watershed. Turbidity is low in these streams, as it is in most Maine rivers and streams (Belknap et al., 1988). The primary source of marine waters to the Taunton Bay is Frenchman Bay, located directly to the south.

All soils mapped in the estuary are in the suborder Aquents because they are permanently saturated (Soil Survey Staff, 2003). The parent materials for the surface soils in the entire estuary are fi ne-silty and fi ne-loamy estuarine sediments (FES; Flannagan and Osher, 2006). In the center of the estu-ary, the FES is >5 m deep. Close to the estuary edges, FES overlies soils that were subaerial before being submerged by the estuary’s rising water level. A 210Pb analysis of three Taunton Bay soil profi les identifi ed that the FES at 30-cm depth are approximately 100 yr old (Osher et al., 2006). This suggests that the most recent inundation and burial of the soils at the edges of the estuary began within the last 100 yr. Cesium-137 counts of the same samples identifi ed minimal disturbance of the soil horizons between 20- and 40-cm depth. Above 20 cm, anomalous 210Pb ages are assumed to be a result of mixing of the surface soils by polychaete worms (Osher et al., 2006).

MATERIALS AND METHODSSoil cores were extracted from the estuary using 7.6-cm-diam-

eter Al pipes. The base of each pipe was fi tted with a core catcher, an Al device designed to allow the sample to slide into the pipe as it is pushed down, but prevent the sample from sliding out of the bottom when the pipe is extracted from the surrounding soil. Each pipe was pushed into the soil surface using a vibracorer (Hoyt and Demarest,

Fig. 1. Map of the Maine coast identifying the four coastal com-partments as delineated by Kelley (1987) with an arrow identifying the location of Taunton Bay within the Island–Bay Complex. The inset map illustrates the location of the Maine coast within the Gulf of Maine Bight.

374 SSSAJ: Volume 71: Number 2 • March –April 2007

1981) mounted on a 6.1-m (20-ft) pontoon boat. Before extraction of the Al pipe containing the soil core, the sampling team measured the distance from the top of the pipe to the soil surface (outside the pipe) and the distance from the top of the pipe to the top of the soil sample. The difference between these two measurements was used to estimate the amount of compaction of the profi le within each pipe. Cores were transported to the University of Maine and stored in a cooler at 4°C. Sampling locations are shown on the landscape unit map in Fig. 2.

In the lab, the Al pipes were split in half using a circular saw. Soil profi les were described from the faces of the split cores according to National Soil Survey Center guidelines (Schoeneberger et al., 2002). Soils were sampled by horizon from one half of the core, dried, sieved to remove course fragments >2 mm, ground using a fl ail sieve (Custom Laboratory Equipment, Orange City, FL), placed in airtight bags, and stored in a cool, dark location. The soil in the other half of the core was covered, returned to the walk-in cooler, and later used to determine the bulk density for each horizon by dividing the horizon volume by the horizon (105°C) oven-dry weight (Blake and Hartge, 1986).

Additional samples of the high-n-value (fl uid) surface horizons were collected from almost all sampling locations 1 yr after the vibra-core samples were collected. The surface soils of the mussel shoals were not collected because these soils do not have a high-n-value surface layer. Original vibracore sampling locations were identifi ed using a global positioning system (GPS) unit. At each location, three samples were collected using a Macaulay sampler. The depth of the high-n-value horizon at each location was measured in the fi eld and

compared with the profi le description for that location. Samples from the top 4 cm were placed in plastic jars, sealed, labeled, transported on ice to the University of Maine-Orono cam-pus, and stored in a refrigerator until analysis. Bulk densities were calculated using the mean weight and calculated volume of samples from the top 4 cm of the soils at each sampling location.

Organic C concentration was deter-mined using a subsample of each horizon that had been dried, sieved, and ground. The subsamples were weighed and encap-sulated in 9- by 10-mm tin cups. Total N and total C were measured using a PDZ Europa (Sandbach, UK) C/N elemental analyzer after combustion to CO2 and N2 at 1000°C. An additional subsample from each horizon was acidifi ed with 0.5 M HCl to remove inorganic C (Midwood and Boutton, 1998). This subsample was rinsed with distilled water, frozen, freeze-dried, weighed, and encapsulated into 9- by 10-mm tins cups as described above. Organic C concentration in these samples was measured on the same elemental ana-lyzer after sample combustion to CO2 and N2 at 1000°C. Inorganic C concentration was determined by calculating the differ-ence between the C concentration before and after acid treatment.

Radiocarbon ages of soil OC were determined on acidifi ed, dried,

ground samples. Each sample was graphitized and analyzed at the W.M. Keck Carbon Cycle Accelerator Mass Spectrometry Laboratory at the University of California, Irvine. Radiocarbon concentrations were corrected for isotopic fractionation and backgrounds were sub-tracted based on measurements of 14C-free coal following the proce-dures of Stuiver and Polach (1977).

Total soil (OC) contents were calculated using the following for-mula (Veldkamp, 1994): SOC = Cs × L × Db × 104, where Cs is the OC concentration (kg C kg−1 soil), L is the soil layer thickness (m), and Db is the bulk density of the soil (g cm−3). The SOC for each horizon to 1 m was calculated by plugging in the depth (L), bulk density (Db), and the C content (Cs) for each horizon into the equation. The soil OC content for each profi le (to 1 m) was calculated as the sum of all horizons in the profi le from 0 to 100 cm.

Groupings of Profi lesSoil profi les were grouped into the landscape units identifi ed by

Flannagan and Osher (2006) in an effort to interpret the role of land-scape position on soil OC contents and distribution. Next, soils were regrouped, taking into account the depth of the FES and presence of another parent material, in addition to location on the landscape. All profi les with <45 cm of FES over some other, low-OC-content parent material were collected from the estuary edge or coastal cove landscape units. These profi les were grouped together and named recently sub-merged edge and cove soils. In the remainder of the profi les from the estu-

Fig. 2. Subaqueous landform map for Taunton Bay, Maine. The map shows seven landscape units as identifi ed by Flannagan and Osher (2006). The locations of soil profi les studied for this research are identifi ed by points and labeled with profi le number.

SSSAJ: Volume 71: Number 2 • March –April 2007 375

ary edges and coastal cove landscape units, FES extended to a depth >90 cm but <200 cm. These profi les were grouped together and named deep edge and cove soils. The four profi les from areas mapped as submerged fl uvial stream or submerged marsh soils (Flannagan and Osher, 2006) were grouped together and given the eponymous name submerged fl uvial stream and marsh soils. The profi les in which FES was the only parent material to a depth of at least 2 m were grouped together and named bay bottom soils. The pairs of soil pro-fi les from the mussel shoals and from the channel shoulder landscape units were placed in their own eponymous groups.

The mean OC content for each group was calculated as the grand mean OC content for the set of profi les within that group. The mean OC content for the entire Taunton Bay estuary (to 1-m depth) was calculated by multiplying the grand mean for each group by the estimated aerial extent of estuary occupied by each group. The OC stored at the base of the channel was not included in this calculation.

Data AnalysisThe weighted mean OC concentration for the top 1 m of soil in

each landscape unit was determined as the product of the OC concentra-tion and the horizon depth for each horizon in each profi le. The sum of the weighted values from each horizon was calculated and divided by the total depth of the profi le to determine a weighted value for each individ-ual profi le. The fi nal value represents the average of the weighted means of profi les within each landscape unit. Statistical analyses were conducted in SYSTAT (Systat Software, 2004). A one-way ANOVA (α = 0.05) was used to test for differences in the mean soil OC contents and mean bulk densities for each landscape unit. The Bonferroni correction factor was used to correct p values for the increased probability of Type I errors.

RESULTSOrganic Carbon Age

Individual radiocarbon ages for samples collected from the horizon spanning the 1-m depth in Profi les 13 and 19 are 3460 ± 15 and 3435 ± 15 yr BP, respectively (Table 1). The radiocarbon measurements for these and the seven other horizons in the FES and at the interface of the FES and the Presumpscot Formation identify a pattern of incremental linear change in OC age (y) in years before present with depth (x) in centimeters (r2 = 0.77).

The primary source of the mineral material in these soils is the silty-clay-textured glaciomarine Presumpscot Formation. The Presumpscot Formation was encountered at the base of some of the soil profi les from the estuary edge and coastal cove landscape units. The mean OC concentration of this material is 0.17 ± 0.15% and the median OC concentration is 0.11%. Radiocarbon ages of the top of the Presumpscot Formation were 5830 ± 20 and 5060 ± 15 yr BP in Profi les 13 and 19, respectively (Table 1).

Organic Carbon ConcentrationWhen profi les were grouped by the landforms identi-

fi ed on Fig. 2, there were no signifi cant differences in soil OC concentration between any of the groups. In addition, the means for the terrestrial edge and coastal cove landscape units had large ranges in soil OC concentration and large standard deviations. The four other landscape unit groups—submerged fl uvial stream, fl uvial marine terrace, mussel shoal (MS) and channel shoulder (CS)—had small ranges in soil OC concen-trations and small standard deviations of the mean.

By regrouping the profi les according to depth of the FES parent material and then according to landscape position, spe-cifi c patterns of OC concentration change with depth were more visible (Fig. 3). The soils of the recently submerged edge and cove (RSEC) group have large decreases in OC concentra-tion from the surface to the underlying horizons (Fig. 3a). The OC concentrations of individual horizons range from close to 4.0% at the surface to as little as 0.11% at depth (Fig. 3a). The mean OC concentrations of RSEC soils are 2.4% at the surface and 0.86% at 1 m. The surface horizons have a large range in OC concentration. The horizons at depths >45 cm have a very small range in OC concentration. The pattern of rapid decrease in OC with depth below the surface is com-monly observed in upland soils.

The submerged fl uvial stream and marshes (SFSM) pro-fi les have the most irregular decreases and increases in OC con-centration from one horizon to the next, as would be expected for soils formed from parent materials deposited by streams and tidal wetlands. Organic C concentrations range from 1.8 to 5.0% (Fig. 3b). The mean OC concentrations are 3.1% at the surface and 1.8% to 100 cm. The range in OC concentra-tion in the surface horizons of the SFSM soils is similar to the surface horizons of the RSEC profi les. At depths >45 cm, the OC concentration range is much greater in the SFSM profi les than in the RSEC profi les.

The profi les in the bay bottom (BB) group have OC con-centrations ranging from 1.7 to 3.3% (Fig. 3d). The mean OC concentrations are 2.5% at the surface and 2.4% to 100 cm. The fi ve soil profi les in the BB group have the least vari-ability in OC concentration from the surface to 1-m depth of sets of soils from any landscape units (Fig. 3). The soils from the MS have OC concentrations ranging from 1.8 to 3.7% (Fig. 3e) and have an average concentration of 2.9%. The two profi les sampled were remarkably similar from the soil surface to approximately 60 cm. At depths >60 cm, however, the dif-ference in OC concentration between the two profi les was as much as 1%.

The weighted mean OC concentration for the top 1 m of soil in the six C groups is presented in Table 2. The mean OC

Table 1. Radiocarbon dates of samples for ten horizons from the Taunton Bay estuary profi les.

Profi le no.

Horizon depth Parent material‡

14C ageUCIAMS

no.§Upper Lower Midpoint†——— cm ———— yr BP

13 78 109 93.5 FES 3435 ± 15 2543513 134 143 137.5 FES & OM 3900 ± 15 2543613 157 175 166.0 PF 5830 ± 20 2543717 97 102 99.5 FES 2285 ± 15 2543819 113 129 126.0 FES 3460 ± 15 2543919 176 180 178.0 OM 4080 ± 20 2544019 180 190 185.0 OM & PF 5060 ± 15 2544120 14 29 21.5 FES 1500 ± 30 2544220 29 40 34.5 FES 1640 ± 15 2544331 96 109 102.5 FES 4890 ± 20 25444

† The midpoint of the horizon was used as the sample depth for the regression analyses.

‡ FES is fi ne estuarine sediment, PF is the Presumpscot Formation, and OM is organic matter.

§ Sample number assigned by the University of California Irvine AMS Laboratory.

376 SSSAJ: Volume 71: Number 2 • March –April 2007

concentration of the soils in the RSEC group is 0.9 ± 0.33%. The RSEC is signifi cantly less than the OC concentration in the rest of the landscape units. The mean OC concentration for the soils of the other groups ranges from 2.0 ± 0.03% in the CS group to 2.9 ± 0.21% in the MS group, but are not signifi -cantly different from one another. The mean OC concentra-tion in all horizons in the top 1 m with FES parent materials is 2.4 ± 0.34%. The concentrations are all close to this mean, fl uctuating only slightly from one horizon to the next.

Bulk DensityThe Db of these estuary soils ranges from fl uid (0.02 g cm−3)

at the soil–water interface, to as much as 2.34 g cm−3 at 100 cm (Fig. 4). The Db of the high-n-value surface layers ranges

between 0.10 and 0.16 g cm−3. This layer is present in all but the MS soils (Fig. 4e). The Db of the surface horizons collected by vibracore range between 0.02 and 0.35 g cm−3. Bulk density increased below these surface horizons in all profi les. The pro-fi les with highest and the greatest range in Db are those in the RSEC group (Fig. 4a). The mean Db of the profi les all other groups is 0.67 ± 0.18 g cm−3. The profi les with the lowest and the smallest range in Db are those in the MS group (0.37–0.85 g cm−3). The mean Db of the RSEC profi les are signifi cantly higher (p < 0.05) than the mean Db in the SFSM, BB, and MS groups (Table 2).

The Db in the BB, SFSM, and deep edge and cove (DEC) profi les (Fig. 4b–4d) increase regularly with depth. Within these three groups of profi les, the range in Db is smallest for the

BB group and greatest for the DEC group. The CS profi les have an irregular pattern of change in subsurface Db with depth.

Total Organic CarbonThe weighted mean OC content for all

estuary soils in Taunton Bay is 136 Mg ha−1. The mean OC content in the RSEC profi les, 67 ± 30 Mg ha−1, is signifi cantly less than the other soil profi le groups. The OC in the RSEC profi les is not equally distributed. Almost all of the OC in the soils is in the top 30 to 45 cm. There is very little OC in the portion of these soils that is below the FES. The mean OC contents of the SFSM and DEC groups are 177 ± 46 and 164 ± 29 Mg ha−1, respectively. These OC contents are signifi cantly greater than the OC contents in all other soil groups (Table 2). The mean OC con-tents in the profi les of the BB, MS, CS, and DEC groups range from 126 ± 6.0 to 147 ± 59 Mg ha−1 and are not signifi cantly different from one another. Variability in OC content between indi-vidual profi les is highest in the SFSM group.

DISCUSSIONThe top 1 m of the majority of the soils

of Taunton Bay are composed of FES, with an average Db of 0.67 g cm−3. The exception is the group of soils referred to as the RSEC. The RSEC soils have FES parent material to approx-imately 20 to 40 cm. Below these fi ne estuarine sediments are submerged upland and coastal landscapes including upland soils, beaches, or fl uvial delta soils. The DEC soils are similar to the RSEC soils in that their profi les (top 2 m) have surface horizons with FES parent material over the inundated and submerged soils that were previously subaerial. The FES layers have Db similar or slightly greater than the other FES horizons in the estuary. The Db of the FES surface is <0.35 g cm−3, and the Db of the FES horizons beneath the soil–water interface are all <1.0 g cm−3 at depth. The horizons buried beneath the FES materials have OC contents and bulk densities similar to the nonsubmerged

Fig. 3. Organic C concentrations for soil horizons from all profi les from each of the six profi le groups: (a) recently submerged edges and coves, (b) submerged fl uvial streams and marshes, (c) deep edge soils, (d) bay bottom, (e) mussel shoal, and (f) channel shoulder. Individual points represent the midpoint of each soil horizon.

SSSAJ: Volume 71: Number 2 • March –April 2007 377

upland soils upslope from the water’s edge (NRCS, 2006). Few of the soils in the RSEC and DEC groups are deeper than 200 cm. Most have bed-rock or glaciomarine sediment within 150 cm of the soil–water interface.

Organic Carbon Content in Taunton Bay Soils vs. Soils of Other Ecosystems

The weighted mean OC con-tent for all estuary soils in Taunton Bay is 136 Mg ha−1, which is 35% to >100% greater than the upland soils that commonly occur within 161 km (100 mi) of Maine’s Island–Bay coast (Table 3). The OC content of these estu-ary soils is similar to, and in some cases greater than, the OC stored in the upland soils of Maine and double the global average soil OC content for Entisols. When compared to freshwater wet-lands, however, the mean OC content for these estuary soils is quite low. The Taunton Bay soils have OC contents equivalent to only half of the OC in Maine’s Fluvaquents, 20% the OC in wetland soils around the globe, and 15% of the OC stored in boreal forest soils (Table 3).

The Taunton Bay soils with FES parent materials to a depth of at least 1 m occupy 78% of the total estuary area, and have OC contents of 126 Mg ha−1 and greater (Table 2). There are soils with FES to 1 m from the edges of the estuary across the entire bay bottom and to the edges of the channels. Their OC contents are greater than the most common off-drained (somewhat poorly, poorly, and very poorly drained) upland soils of Downeast Maine (Table 3). This category includes two groups of soils with signifi cantly higher OC contents: the SFSM soils (177 Mg ha−1) and soils at the edges of the estuary and in coves (DEC, 164 Mg ha−1).

The soils of the SFSM and DEC have mean OC concentrations similar to the soils of the BB, but have greater Db (Table 2). It is the increased Db that results in their OC content being sig-nifi cantly higher than the other soils. Before becoming submerged by the saline estuary waters, the SFSM soils would have been mapped as Fluvaquents (Flannagan and Osher, 2006). As such, it is not a surprise that their OC contents are similar to the Fluvaquents (regularly fl ooded, nonsaline soils) mapped throughout Maine.

The RSEC soils have the lowest OC con-tents of all Taunton Bay estuary soils (Table 2). These are the only soils in the estuary with OC contents equivalent to the global average for the Entisol soil order. They occupy approximately 7% of the estuary, and have OC contents similar

Table 2. Area of each profi le group, number of profi les described in each group, mean organic carbon content and mean bulk density for the six groups of soil profi les.

Name of profi le group Area† n‡Organic C§

Bulk density§Content Conc.

% no. Mg ha−1 (SD) % (SD) g cm−3 (SD)Recently submerged edges & coves 7 5 67 a¶ (30) 0.9 a (0.33) 1.13 a (0.21)Submerged fl uvial streams & marshes 3 4 177 b (46) 2.6 b (0.47) 0.70 b (0.10)Deep edge & cove 6 4 164 b (29) 2.3 b (0.43) 0.81ab (0.23)Bay bottom 51 5 134 ab (34) 2.4 b (0.28) 0.57 b (0.07)Mussel shoal 1 2 126 ab (6) 2.9 b (0.21) 0.46 b (0.04)Channel shoulder 17 2 147 ab (59) 2.0 b (0.34) 0.80 ab (0.21)

† Estuary area occupied by soils of this group based on Flannagan and Osher (2006). The sum of these areas is 85%, the channel occupies the other 15% of the estuary.

‡ Number of profi les used to calculate mean organic C and bulk density for profi le groups.

§ Means for top 1 m of soil.

¶ Means in the same column with the same letter indicate no signifi cant difference at α = 0.05.

Fig. 4. Bulk density for soil horizons from all profi les from each of the six soil profi le groups: (a) recently submerged edges and coves, (b) submerged fl uvial streams and marshes, (c) deep edge soils, (d) bay bottom, (e) mussel shoal, and (f) chan-nel shoulder. individual points represent the midpoint of each soil horizon.

378 SSSAJ: Volume 71: Number 2 • March –April 2007

to Colonel and Monadnock soils—well-drained soils formed from glacial till.

The mean OC concentrations of all soils in Taunton Bay are similar to near-shore marine sediments in Massachusetts and in the same range as OC in subaqueous soils of coastal lagoons in Maryland and Rhode Island (Table 3).

CONCLUSIONSThis research is the fi rst to quantify the OC stored to

a depth of 1 m in estuary soils not dominated by marsh or mangrove vegetation. Because it is a fi rst step, we made sure to begin this research by quantifying the range in soil OC across the entire landscape, and identifying differences in OC content between soils with varying parent materials and landscape posi-tion. As more estuary OC inventories are completed, it will be possible to scale up the results to regional or global estimates of soil OC in non-marsh estuarine ecosystems.

ACKNOWLEDGMENTSNational Research Initiative Competitive Grant no. 2003-35107-13885 from the USDA Cooperative State Research, Education, and Extension Service supported this project, as did ME Sea Grant, the Maine Agricultural and Forest Experiment Station, and a University of Maine Faculty Equipment Grant. Nick Brown, director of the University of Maine’s Center for Cooperative Aquaculture Research, provided logistics including a local base of operations. Christopher Flannagan collected and described the soil profi les used in this research. Our thanks go to the University of Maine undergraduate students who assisted in the fi eld and laboratory: John Boucher (2004), Patricia Rouleau (2005), and Erica Davis (2006).

REFERENCESAnderson, F.E., L. Black, L.E. Watling, W. Mook, and L.M. Mayer. 1981. A temporal and spatial study of mudfl at erosion and deposition. J. Sediment. Petrol. 51:0729–0736.Balduff, D.M., and M.C. Rabenhorst. 2006. Subaqueous landforms and soils of Chincoteague Bay, Maryland, USA. In Proc. World Congr. Soil Sci., 18th, Philadelphia, PA. 9–15 July 2006.Belknap, D.F., J.T. Kelley, and D.H. Robbins. 1988. Sediment dynamics of the near shore Gulf of Maine: Submersible experimentation and remote sensing. p. 143–176. In I. Babb and M. De Luca (ed.) Benthic productivity and marine resources of the Gulf of Maine. NOAA Natl. Undersea Res. Progr., Res. Rep. 88-3. Natl. Oceanic Atmos. Admin., Silver Spring, MD.Blake, G.R., and K.H. Hartge. 1986. Bulk density. p. 363–382. In A. Klute (ed.) Methods of soil analysis. Part 1. Physical and mineralogical methods. Agron. Monogr. 9. 2nd ed. ASA and SSSA, Madison, WI.Blazejewski, G.A., M.H. Stolt, A.J. Gold, and P.M. Groffman. 2005. Macro and micromorphology of subsurface carbon in riparian zone soils. Soil Sci. Soc. Am. J. 69:1320–1329.Bloom, A.L. 1960. Late Pleistocene changes of sea level in southwestern Maine. Maine Geol. Surv., Augusta.Bradley, M.P., and M.H. Stolt. 2003. Subaqueous soil–landscape relationships in a Rhode Island estuary. Soil Sci. Soc. Am. J. 67:1487–1497.

Brady, N.C., and R.R. Weil. 1996. The nature and properties of soils. 11th ed. Prentice Hall, Upper Saddle River, NJ.

Chmura, G.L., S.C. Anisfeld, D.R. Cahoon, and J.C. Lynch. 2003. Global carbon sequestration in tidal, saline wetland soils. Global Biogeochem. Cycles 17:1111, doi:10.1029/2002GB001917.

Choi, Y., and Y. Wang. 2004. Dynamics of carbon sequestration in a coastal wetland using radiocarbon measurements. Global Biogeochem. Cycles 18(4):GB4016, doi:10.1029/2004GB002261.

Coppock, C., and M. Rabenhorst. 2003. Subaqueous soils of Rehoboth Bay, DE: Soil mapping in a coastal lagoon. In Agronomy abstracts [CD-ROM]. ASA, Madison, WI.

Davidson, E.A., and P.A. Lefebvre. 1993. Estimating regional carbon stocks and spatially covarying edaphic factors using soil maps at three scales. Biogeochemistry 22:107–131.

Demas, G.P., and M.C. Rabenhorst. 1999. Subaqueous soils: Pedogenesis in a submersed environment. Soil Sci. Soc. Am. J. 63:1250–1257.

Demas, G.P., M.C. Rabenhorst, and J.C. Stevenson. 1996. Subaqueous soils: A pedological approach to the study of shallow-water habitats. Estuaries 19:229–237.

Dorion, C.C., G.A. Balco, M.R. Kaplan, K.J. Kreutz, J.D. Wright, and H.W. Borns, Jr. 2001. Stratigraphy, paleoceanography, chronology, and environment during deglaciation of eastern Maine. Geol. Soc. Am. Spec. Pap. 351:215–242.

Fischler, K., L. Ellis, and M. Collins. 2005. Landscape modeling of a spoil island: Applications of pedology to subaqueous soils and seagrass mitigation in the Indian River Lagoon. In Agronomy abstracts [CD-ROM]. ASA, Madison, WI.

Flannagan, C.F., and L.J. Osher. 2006. Soil–landform relationships in shallow estuarine ecosystems of Downeast Maine. In Proc. World Congr. Soil Sci., 18th, Philadelphia, PA. 9–15 July 2006.

Folger, D.W. 1972. Texture and organic carbon content of bottom sediments in some estuaries of the United States. Geol. Soc. Am. Mem. 133:391–408.

Fung, I.Y., S.C. Doney, K. Lindsay, and J. John. 2005. Evolution of carbon sinks in a changing climate. Proc. Natl. Acad. Sci. 102:11201–11206.

Table 3. Organic C concentrations and contents to the 1-m depth for a range of soils and sediments compared with soils from Taunton Bay estuary.

Description Location Organic C† Reference

% Mg ha−1

Upland soils

Colonel (well drained) Maine 1.3 63 Rourke (1994)

Monadnock (well drained) Maine 1.6 75 Rourke (1990)

Nicholville (mod. well drained) Maine 0.2 15 Rourke (1990)

Scantic (poorly drained) Maine 0.9 101 Rourke and Schmidt (1979)

Burnham (v. poorly drained) Maine 0.8 91 Rourke (1990)

Fluvaquents (v. poorly drained) Maine average 1.1 260 Davidson and Lefebvre (1993)

Entisols Global average 69 Kern (1994)

Wetland soils Global average 720 U.S. Dep. of Energy (1999)

Boreal forest wetland soils Manitoba, CA 1000 Trumbore and Harden (1997)

Estuary soils and sediments

Submerged edges and coves Maine 1.1 67 this study

Bay bottom Maine 2.6 134 this study

Channel shoulder Maine 2.4 147 this study

Submerged streams and marshes Maine 3.2 177 this study

Weighted mean, entire estuary Maine 1.1–3.2 136 this study

Estuarine sediment Massachusetts 1.8–2.6 Latimer et al. (2003)

Coastal lagoon Maryland 0–1.2 Demas et al. (1996)

Coastal lagoon Rhode Island 1.0–4.0 Bradley and Stolt (2003)

† Organic C values for Maine upland soils are weighted means for the top 1 m of representative soil profi les.

SSSAJ: Volume 71: Number 2 • March –April 2007 379

Gacia, E., and C.M. Duarte. 2001. Sediment retention by a Mediterranean Posidonia oceanica meadow: The balance between deposition and resuspension. Estuarine Coastal Shelf Sci. 52:505–514.

Gacia, E., T.C. Granata, and C.M. Duarte. 1999. An approach to measurement of particle fl ux and sediment retention within seagrass (Posidonia oceanica) meadows. Aquat. Bot. 65:255–268.

Gaudinski, J.B., S.E. Trumbore, E.A. Davidson, and S. Zheng. 2000. Soil carbon cycling in a temperate forest: Radiocarbon-based estimates of residence times, sequestration rates and partitioning of fl uxes. Biogeochemistry 51:33–69.

Gehrels, W.R., D.F. Belknap, S. Black, and R.M. Newnham. 2002. Rapid sea-level rise in the Gulf of Maine, USA, since AD 1800. Holocene 12:383–389.

Guo, Y., R. Amundson, P. Gong, and Q. Yu. 2006. Quantity and spatial variability of soil carbon in the conterminous United States. Soil Sci. Soc. Am. J. 70:590–600.

Houghton, R.A. 1995. Change in the storage of terrestrial carbon since 1850. In R. Lal et al. (ed.) Soils and global change. CRC Press, Boca Raton, FL.

Houghton, R.A. 2003. Why are estimates of the terrestrial carbon balance so different? Global Change Biol. 9:500–509.

Hoyt, W.H., and J.M. Demarest. 1981. A versatile twin-hull barge for shallow-water vibracoring. J. Sediment. Res. 51:656–657.

Hurtt, G.C., S.W. Pacala, P.R. Moorecroft, J. Caspersen, E. Shevialkova, R.A. Houghton, and B. Moore III. 2002. Predicting the future of the U.S. carbon sink. Proc. Natl. Acad. Sci. 99:1389–1394.

Hussein, A.H., M.C. Rabenhorst, and M.L. Tucker. 2004. Modeling of carbon sequestration in coastal marsh soils. Soil Sci. Soc. Am. J. 68:1786–1795.

Jacobson, H.A., J.L. Jacobson, and J.T. Kelley. 1987. Distribution and abundance of tidal marshes along the coast of Maine. Estuaries 10:126–131.

Jordan, G.B. 1999. Soil survey of Hancock County area, Maine. Nat. Resour. Conserv. Serv., Washington, DC.

Kelley, J.T. 1987. An inventory of coastal environments and classifi cation of Maine’s glaciated shoreline. p. 364. In D.M. Fitzgerald and P.S. Rosen (ed.) Glaciated coasts. Academic Press, San Diego.

Kern, J.S. 1994. Spatial patterns of soil organic carbon in the contiguous United States. Soil Sci. Soc. Am. J. 58:439–455.

Knorr, W., I.C. Prentice, J.I. House, and E.A. Holland. 2005. Long term sensitivity of soil carbon turnover to warming. Nature 433:298–301.

Lacelle, B., S. Waltman, N. Bliss, and F. Orozco-Chavez. 2001. Methods used to create the North American soil organic carbon digital database. p. 485–494. In R. Lal et al. (ed.) Assessment methods for soil carbon. Adv. Soil. Sci. CRC Press, Boca Raton, FL.

Larsen, P., L. Doggett, C. Garside Topinka, J. Mague, T.T. Garfi eld, R. Gerber, S.I. Fefer, P. Schettig, and L. Thornton. 1980. The estuarine system. p. 5-1–5-148. In S.I. Fefer and P.A. Schettig (ed.) An ecological characterization of coastal Maine (north and east of Cape Elizabeth). Vol. 2. U.S. Fish and Wildl. Serv., Newton Corner, MA.

Latimer, J.S., W.S. Boothman, C.E. Pesch, G.L. Chmura, V. Pospelova, and S. Jayaraman. 2003. Environmental stress and recovery: The geochemical record of human disturbance in New Bedford Harbor, Apponagansett Bay, Massachusetts (USA). Sci. Total Environ. 313:153–176.

Matthews, H.D., A.J. Weaver, and K.J. Meisnner. 2005. Terrestrial carbon cycle dynamics under recent and future climate change. J. Clim. 18:1609–1628.

Midwood, A.J., and T.W. Boutton. 1998. Soil carbonate decomposition by acid has little effect on δ13C of organic matter. Soil Biol. Biochem. 30:1301–1307.

New York City Soil Survey Staff. 2005. Reconnaissance soil survey. Nat. Resour. Conserv. Serv., Staten Island.

NRCS. 2006. Web soil survey. Available at websoilsurvey.nrcs.usda.gov/app/ (accessed 17 Mar. 2006; verifi ed 5 Sept. 2006).

Odum, E.P. 1959. Fundamentals of ecology. W.S. Saunders Co., Philadelphia, PA.Osher, L.J., L. Leclerc, G.B. Wiersma, C.T. Hess, and V.E. Guiseppe. 2006. Heavy

metal contamination from historic mining in upland soil and estuarine sediments of Egypt Bay, Maine, USA. Estuar. Coastal Shelf Sci. (in press).

Pacala, S., G. Hurtt, D. Baker, P. Devlin, R.A. Houghton, R.A. Birdsley et al. 2001. Consistent land and atmosphere based U.S. carbon sink estimates.

Science 292:2316–2320.Payne, M.K., and M.H. Stolt. 2006. Subaqueous soils, water quality, and

estuary health. In Proc. World Congr. Soil Sci, 18th, Philadelphia, PA. 9–15 July 2006.

Potsma, H. 1967. Sediment transport and sedimentation in the estuarine environment. p. 158–184. In G.H. Lauff (ed.) Estuaries. Am. Assoc. Adv. Sci., Washington, DC.

Prentice, I.C., G.D. Farquahr, M.J.R. Fasham, M.L. Goulden, M. Heimann, V.J. Jaramillo, H.S. Kheshgi, C. Le Querere, and R.J. Scholes. 2001. The carbon cycle and atmospheric carbon dioxide. p. 183–237. In J.T. Houghton et al. (ed.) Climate change 2001: The scientifi c basis. Cambridge Univ. Press, Cambridge, UK.

Raymond, P.A., and J.E. Bauer. 2001. Use of 14C and 13C natural abundances for evaluating riverine, estuarine, and coastal DOC and POC sources and cycling: A review and synthesis. Org. Geochem. 32:469–485.

Rourke, R.V. 1990. Chemical and physical properties of the Aurele, Burnham, Easton, Lille, Linneus, Madonadnock, Nicholville, and Tunbridge soil map units. Maine Agric. For. Exp. Stn. Tech. Bull. 137. Univ. of Maine, Orono.

Rourke, R.V. 1994. Chemical and physical properties of the Chesuncook, Colonel, Dixfi eld, and Telos soil map units. Maine Agric. For. Exp. Stn. Tech. Bull. 155. Univ. of Maine, Orono.

Rourke, R.V., and K.A. Schmidt. 1979. Chemical and physical properties of the Boothbay, Brayton, Croghan, Mondarda, Plaisted, Scantic, and Swanville soil mapping units. Life Sci. Agric. Exp. Stn . Tech. Bull. 94. Univ. of Maine, Orono.

Saunders, T.J., and M. Collins. 2006. Quantifying the biogeochemical function of the subaqueous soils in the environment. In Proc. World Congr. Soil Sci., 18th, Philadelphia, PA. 9–15 July 2006.

Schlesinger, W.H. 1986. Changes in soil carbon storage and associated properties with disturbance and recovery. p. 194–220. In J.R. Trebalka and D.E. Reichle (ed.) The changing carbon cycle—A global analysis. Springer-Verlag, New York.

Schnitker, D. 1974. Postglacial emergence of the Gulf of Maine. Geol. Soc. Am. Bull. 85:4491–4494.

Schoeneberger, P.J., D.A. Wysocki, E.C. Benham, and W.D. Broderson. 2002. Field book for describing and sampling soils. 2nd ed. Natl. Soil Surv. Center, Lincoln, NE.

Smith, S.V., W.H. Renwick, R.W. Buddemeier, and C.J. Crossland. 2001. Budgets of soil erosion and deposition for sediments and sedimentary organic carbon across the conterminous United States. Global Biogeochem. Cycles 15:697–707.

Soil Survey Staff. 2003. Keys to Soil Taxonomy. 9th ed. Nat. Resour. Conserv. Serv., Washington, DC.

Stuiver, M. and H.A. Polach. 1977. Discussion: Reporting of 14C data. Radiocarbon 19: 355–363.

Systat Software. 2004. SYSTAT 11. Statistics for 1, 2, 3 for Windows. Systat Software, Point Richmond, CA.

Thom, R.M., A.B. Borde, G.D. Williams, D.L. Woodruff, and J.A. Southard. 2003. Climate change and seagrasses: Climate-linked dynamics, carbon limitation and carbon sequestration. Gulf Mex. Sci. 21:134.

Thompson, J.A., and R.K. Kolka. 2005. Soil carbon storage estimation in a forested watershed using quantitative soil-landscape modeling. Soil Sci. Soc. Am. J. 69:1086–1093.

Trumbore, S.E., and J.W. Harden. 1997. Accumulation and turnover of carbon in organic and mineral soils of the BOREAS northern study area. J. Geophys. Res. 102:28817–28830.

U.S. Department of Energy. 1999. Carbon sequestration: State of the science. Carbon sequestration research and development. U.S. Dep. of Energy, Offi ce of Fossil Energy, Washington, DC.

Veldkamp, E. 1994. Organic carbon turnover in three tropical soils under pasture after deforestation. Soil Sci. Soc. Am. J. 58:175–180.

Wang, Y., and Y. Hseih. 2002. Uncertainties and novel prospects in the study of the soil carbon dynamics. Chemosphere 49:791–804.